WO2009125479A1

WO2009125479A1 - Magnetic pole position detector and method for detecting magnetic pole position

Info

Publication number: WO2009125479A1
Application number: PCT/JP2008/057006
Authority: WO
Inventors: 丈昌荒川; 修也佐野
Original assignee: 三菱電機株式会社
Priority date: 2008-04-09
Filing date: 2008-04-09
Publication date: 2009-10-15
Also published as: DE112008003786T5; US20110031962A1; TW200943691A; JPWO2009125479A1; CN101990733B; JP4889808B2; KR20100133388A; TWI361554B; US8339125B2; KR101135733B1; CN101990733A

Abstract

A magnetic pole position detector which is capable of acquiring a magnetic pole position properly even when two magnetic sensors are arranged at an arbitrary spacing interval, wherein each sensor output of two magnetic sensors arranged at a given spacing interval L along the magnetic pole array direction of a motor is a and b, while a phase difference φ between two sensor outputs a and b is an input. In case sin φ > δ, tan θ is determined by dividing a*sin φ by b-a*cos φ to output the magnetic pole position θ = tan^-1{a*sin φ/(b-a*cos φ)}. In case sin φ =< δ, the magnetic pole position θ = {sin^-1(a)+cos^-1(b)-φ}/2 is output based on a = sin θ and b = cos(θ + φ). Based on φ = 2π x L/(magnetic pole pitch or number of poles), use of φ as a correction coefficient for two sensor outputs a and b can cancel any magnetic pole position error even when the spacing interval L differs from a theoretical value. That is to say, two magnetic sensors can be arranged at an arbitrary spacing interval.

Description

Magnetic pole position detection apparatus and method

The present invention relates to a magnetic pole position detection apparatus and method for detecting a magnetic pole position where a movable part of a motor is located.

The magnetic pole structure of a motor is a structure in which a plurality of magnets having a certain length (magnetic pole pitch) are linearly arranged with different polarities in a linear motor. In a rotary motor (such as a brushless motor), a predetermined pole is used. In this structure, a predetermined number of magnetic poles having different polarities are annularly arranged so as to form a number.

In a linear motor or a rotary motor (such as a brushless motor), in order to detect magnetism at the magnetic pole position where the movable part is located, two or more magnetic sensors that detect a change amount according to the distance of the magnetic flux generated by the magnetic pole are motors. Are arranged at certain intervals along the magnetic pole arrangement direction.

The magnetic pole position detection device is a device that calculates and detects the magnetic pole position where the movable part is located based on the phase difference between the sensor outputs of two magnetic sensors among the two or more magnetic sensors described above.

The linear motor is equipped with a magnetic pole position detection device in order to obtain position information on which position (magnetic pole position) within the magnetic pole pitch when moving in the magnet arrangement direction. That is, the magnetic pole position detection apparatus used for the linear motor includes two magnetic sensors in the apparatus. On the other hand, in the rotary motor, two or more magnetic sensors are a part of the constituent parts of the motor, and the magnetic pole position detecting device is provided outside the motor.

By the way, regarding the arrangement interval of the two magnetic sensors, conventionally, the two magnetic sensors are arranged such that the phase difference between the sensor outputs is 90 degrees (for example, Patent Documents 1 and 2).

Then, in the example of the linear motor, between the arrangement interval L of the two magnetic sensors and the magnetic pole pitch PIT,
L = PIT / 4 (1)
The relationship holds. The arrangement interval L is represented by an electrical angle in the rotary motor,
L = 180 ° / number of poles.

Since the output waveform of each magnetic sensor has a sine wave shape, when two magnetic sensors are arranged so as to satisfy the constraint condition of this equation (1), the sensor output on the leading phase side is a, and the phase on the lagging phase side The sensor output a can be regarded as a sine wave, and the sensor output b can be regarded as a cosine wave. The magnetic pole position can be obtained by calculating the tangent value of the sensor outputs a and b.

That is, if the magnetic pole position is θ, tan θ = a / b, so the magnetic pole position θ is
θ = tan ⁻¹ (a / b) (2)
It can be asked.

The processing block in this case, that is, the conventional magnetic pole position detection device includes a divider for obtaining a / b, a tan ⁻¹ angle calculator for obtaining an arctangent value from a tangent value tanθ obtained by the divider, and Consists of.

JP-A-60-180468 (FIG. 3) JP 2001-78392 A (FIG. 1)

However, the arrangement interval between the two magnetic sensors in the conventional magnetic pole position detection device cannot be determined arbitrarily, and as can be understood from the constraint condition of Equation (1), the linear motor is determined by the magnetic pole pitch. There is a restriction that the rotational motor is determined by the number of poles.

Therefore, if there is an error in the mounting position of the magnetic sensor, troublesome correction processing is required. And in the magnetic pole position detection apparatus mounted in a linear motor, since the magnitude | size of the said apparatus is decided by magnetic pole pitch, size reduction is difficult. In addition, the magnetic pole position detection device cannot be shared between linear motors having different magnetic pole pitches.

The present invention has been made in view of the above, and an object of the present invention is to provide a magnetic pole position detection apparatus and method that can correctly acquire a magnetic pole position even if two magnetic sensors are arranged at an arbitrary interval.

In order to achieve the above-described object, a magnetic pole position detection device according to the present invention is a sensor of two magnetic sensors among two or more magnetic sensors arranged at a predetermined interval along the magnetic pole arrangement direction of the motor. A magnetic pole position detection device that calculates and detects the magnetic pole position where the movable part is located based on the phase difference between the outputs, wherein the advancing sensor output of the two sensor outputs is a and the lagging sensor output is b When the phase difference is φ, the magnetic pole position θ when the sine value of the phase difference is larger than the determination threshold is θ == tan ⁻¹ {(a · sinφ / (ba · cosφ)} And the magnetic pole position θ when the sine value of the phase difference is smaller than the determination threshold, θ = {sin ⁻¹ (a) + sin ⁻¹ (b) −φ } / 2 and a second arithmetic system for detecting by performing the arithmetic operation It is characterized by.

According to the present invention, it is possible to obtain a magnetic pole position detecting device that can correctly acquire the magnetic pole position even if two magnetic sensors are arranged at an arbitrary interval.

FIG. 1 is a block diagram showing a configuration of a magnetic pole position detection apparatus according to an embodiment of the present invention. FIG. 2 is a diagram for explaining the principle of magnetic pole position detection according to the present invention of the magnetic pole position detection apparatus mounted on the linear motor. FIG. 3 is a waveform diagram showing output waveforms of two magnetic sensors provided in the magnetic pole position detection device shown in FIG. FIG. 4 is a diagram for explaining the magnetic pole position detection operation when the arrangement interval is 5 mm and the magnetic pole pitch is 30 mm in the configuration shown in FIG. FIG. 5 is a diagram showing a simulation result under the conditions shown in FIG. 4 in comparison with the conventional method. FIG. 6 is a diagram for explaining the magnetic pole position detection operation when the arrangement interval is 5 mm and the magnetic pole pitch is 20 mm in the configuration shown in FIG. FIG. 7 is a diagram showing a simulation result under the conditions shown in FIG. 6 in comparison with the conventional method. FIG. 8 is a diagram showing a simulation result of the magnetic pole position obtained by calculating the correction coefficient when the arrangement interval is deviated from the theoretical value in the configuration shown in FIG. 2 in comparison with the conventional method. FIG. 9 is a diagram for explaining a method of performing magnetic pole position detection using three magnetic sensors.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Magnetic pole position detection apparatus 2 Linear motor 3 Guide 4

Magnet

11, 12, 13

Magnetic sensor

20, 21, 31

Multiplier

22, 23, 32

Selection circuit

24, 29 Subtractor 25

Divider

26, 27, 28 Angle calculator 30 Adder a, b Sensor output φ Phase difference (Correction coefficient)

DETAILED DESCRIPTION Exemplary embodiments of a magnetic pole position detection apparatus and method according to the present invention will be described below in detail with reference to the drawings.

FIG. 1 is a block diagram showing a configuration of a magnetic pole position detection device according to an embodiment of the present invention. In this embodiment, the description will be made assuming that the linear motor is used. However, the linear motor can also be used.

As shown in FIG. 1, the magnetic pole position detection apparatus according to this embodiment receives the sensor outputs a and b of the two magnetic sensors and the phase difference φ between the sensor outputs a and b and inputs the magnetic pole position based on them. It is a device for calculating and outputting θ.

1 includes

multipliers

20 and 21,

selection circuits

22 and 23, a subtractor 24, a divider 25,

angle calculators

26, 27, and 28, a subtractor 29, an adder 30, and a multiplier. 31 and a selection circuit 32 are provided. Although not shown, a phase difference calculation unit for obtaining the phase difference φ by the following equation (3) is provided.

This magnetic pole position detection device is configured based on the following principle. This will be described with reference to FIGS. FIG. 2 is a diagram for explaining the principle of magnetic pole position detection according to the present invention of the magnetic pole position detection device mounted on the linear motor. FIG. 3 is a waveform diagram showing output waveforms of two magnetic sensors provided in the magnetic pole position detection device shown in FIG.

2, the linear motor 2 on which the magnetic pole position detection device 1 is mounted can move in the arrangement direction of the magnets 4 having a certain length (magnetic pole pitch) guided by the guide 3. In the magnetic pole position detection device 1, two

magnetic sensors

11 and 12 are arranged at an interval L in the arrangement direction of the magnets 4. Therefore, the output values of the

magnetic sensors

11 and 12 when the magnetic pole position detection device 1 is located at a certain position are different because the distance from the N pole is different.

When the magnetic pole position detection apparatus 1 moves along the guide 3, the outputs (sensor outputs) a and b of the

magnetic sensors

11 and 12 are constant according to the movement position of the magnetic pole position detection apparatus 1 as shown in FIG. It becomes a sine wave shape having a phase difference φ [rad]. The phase difference φ between the sensor outputs a and b is determined by the magnetic pole PIT and the arrangement interval L between the

magnetic sensors

11 and 12,
φ = 2π × L / PIT [rad] (3)
It can be expressed.

Here, if the sensor output a is the output of the magnetic sensor 11 arranged on the phase advance side, and the sensor output b is the output of the magnetic sensor 12 arranged on the slow phase side, the output of the magnetic sensor 11 is assumed. The phase of the output b of the magnetic sensor 12 is delayed by φ = 2π × L / PIT [rad] compared to a. When the magnetic pole position at this time is θ, and the tangent value tan θ = a / b is obtained,
a / b = sin θ / sin (θ + φ) = a / (a · cos φ + cos θ sin φ)
It becomes. Therefore, the cosine value cos θ of the magnetic pole position θ is
cos θ = (ba · cos φ) / sin φ (4)
Can be calculated.

In the conventional magnetic pole position detection method, since the outputs a and b of the

magnetic sensors

11 and 12 have a phase difference of 90 degrees, φ = 90 °, and when applied to the equation (4), cos θ = b Become. Thus, Equation (4) is an extension of the conventional method.

When the magnetic pole position θ is calculated from the tangent value tan θ as in the conventional method,
Since tan θ = sin θ / cos θ = a / cos θ = a · sin φ / (ba−cos φ), the magnetic pole position θ is
θ = tan ⁻¹ {(a · sinφ / (ba−cosφ)} (5)
Can be calculated.

Formula (5) uses the phase difference φ obtained from the magnetic pole pitch and the arrangement interval of the

magnetic sensors

11 and 12 as a correction coefficient for the sensor outputs a and b, and calculates the tangent value tan θ of the magnetic pole position from the sensor outputs a and b. This indicates that the magnetic pole position θ can be obtained.

However, in the equation (5), under the condition that sin φ = 0, tan θ = 0, so cos θ cannot be obtained from the equation (4). Actually, when sinφ << 1, the arrangement interval L of the

magnetic sensors

11 and 12 is extremely small compared to the magnetic pole pitch PIT, and the values of the sensor outputs a and b of the

magnetic sensors

11 and 12 substantially match each other. It becomes difficult to obtain the magnetic pole position θ by obtaining tan θ from 5). Further, when the arrangement interval L of the

magnetic sensors

11 and 12 is an integral multiple of the magnetic pole pitch PIT, sin φ = 0 is similarly obtained, and it is difficult to obtain the magnetic pole position θ by obtaining tan θ from the equation (5). .

Therefore, in such a case, the magnetic pole position θ is calculated by directly obtaining the value sin θ from the sensor outputs a and b. That is, from the relationship of a = sin θ and b = sin (θ + φ),
θ = sin ⁻¹ (a)
θ = sin ⁻¹ (b) −φ
Therefore, in order to calculate the magnetic pole position θ using the two sensor outputs a and b,
θ = {sin ⁻¹ (a) + sin ⁻¹ (b) −φ} / 2 (6)
And deformed. In this way, when sin φ is small, the magnetic pole position θ can be calculated by calculating sin θ directly from the sensor outputs a and b.

In the case of a rotary motor, the magnetic pole position can be calculated in the same procedure as described above by considering the magnetic pole pitch PIT as the number of poles and considering the arrangement interval L, the phase difference φ, and the magnetic pole position θ as electrical angles. .

[Correction based on Rule 91 14.11.2008]
FIG. 1 shows the above contents organized in the form of processing blocks. In FIG. 1, the value δ is calculated by calculating tan θ by the equation (5) according to the magnitude of sin φ = sin (2π × L / PIT), or calculating the magnetic pole position θ by the equation (6). This is a determination threshold value for determining whether to calculate the magnetic pole position θ. This determination threshold δ is a design value determined within a range of 0 <δ << 1.

The multiplier 20 outputs “a · sin φ” obtained by multiplying the sensor output a by sin φ to one input terminal 15 a of the selection circuit 22. The multiplier 21 outputs “a · cos φ” obtained by multiplying the sensor output a by cos φ to one input terminal 16 a of the selection circuit 23. The sensor output a is directly input to the other input terminal 15 b of the selection circuit 22. The value 0 is input to the other input terminal 16 b of the selection circuit 23.

The output of the selection circuit 22 is input to one input terminal of the divider 25 and the angle calculator 27. The output of the selection circuit 23 is input to one input terminal of the subtractor 24. The subtracter 24 outputs a value obtained by subtracting the output of the selection circuit 23 from the sensor output b which is the other input to the other input terminal of the divider 25 and the angle calculator 28.

The angle calculator 26 performs an angle calculation of “tan ⁻¹ ” on the output “tan θ” of the divider 25 to obtain an arc tangent value (magnetic pole position) θ and outputs it to one input terminal 17 a of the selection circuit 32. .

The angle calculator 28 performs an angle calculation of “sin ⁻¹ ” on the output of the subtractor 24, and outputs the obtained inverse sine value to one input terminal of the subtractor 29. The subtractor 29 outputs a value obtained by subtracting the phase difference φ, which is the other input, from the output of the angle calculator 28 to one input terminal of the adder 30.

The angle calculator 27 performs an angle calculation of “sin ⁻¹ ” on the output of the selection circuit 22 and outputs the obtained inverse sine value to the other input terminal of the adder 30. The multiplier 31 outputs a value obtained by multiplying “(output of the subtractor 29) + (output of the angle calculator 27)” output from the adder 30 by ½ to the other input terminal 17 b of the selection circuit 32. .

In the above configuration, each of the

selection circuits

22, 23, and 32 selects one

input terminal

15a, 16a, or 17a when sin φ> δ, and the other input terminal 15b when sin φ ≦ δ. , 16b, 17b are selected.

Then, when sinφ> δ, the output of the selection circuit 22 is a · sinφ, and the output of the selection circuit 23 is a · cosφ. The output of the subtractor 24 is ba · cos φ. The output of the divider 25 is a · sin φ / (ba−cos φ) = tan θ. Therefore, the magnetic pole position θ obtained by the calculation of Expression (5) is obtained from the output of the angle calculator 26.

On the other hand, when sin φ ≦ δ, the output of the selection circuit 22 is the sensor output a, and the output of the selection circuit 23 is the value 0. The output of the subtracter 24 is a sensor output b. The output of the angle calculator 27 is θ = sin ⁻¹ (a). The output of the angle calculator 28 is θ = sin ⁻¹ (b). The output of the subtractor 29 is θ = sin ⁻¹ (b) −φ. The output of the adder 30 is 2θ = {sin ⁻¹ (a) + sin ⁻¹ (b) −φ}. Therefore, the magnetic pole position θ obtained by the calculation of Expression (6) is obtained from the output of the multiplier 31.

1 that the magnetic pole position θ can be calculated at an arbitrary arrangement interval L. The configuration shown in FIG. Even when the arrangement interval L deviates from the theoretical value in the attachment of the

magnetic sensors

11 and 12, by using the phase difference φ as a correction coefficient, the arrangement interval L resulting from the attachment can be reduced without performing a special correction process. The error can be corrected.

That is, when the arrangement interval L deviates from the theoretical value in the attachment of the

magnetic sensors

11 and 12, the actual arrangement interval L is actually measured, or the error of the attachment position is calculated by observing the magnetic pole position error waveform. The actual value of the arrangement interval L is obtained by estimating the arrangement interval L, and the phase difference φ obtained by applying it to the equation (3) may be applied to φ in FIG. 1 as a correction coefficient. That is, even if there is an error in the arrangement interval L, the correct magnetic pole position can be automatically obtained if the correction coefficient φ is appropriately obtained as described above and applied to the configuration of FIG. Hereinafter, various specific embodiments will be described as examples.

Example 1
FIG. 4 is a diagram for explaining the magnetic pole position detection operation when the arrangement interval is 5 mm and the magnetic pole pitch is 30 mm in the configuration shown in FIG. FIG. 5 is a diagram showing a simulation result under the conditions shown in FIG. 4 in comparison with the conventional method.

The correction coefficient φ when the arrangement interval L is 5 mm and the magnetic pole pitch PIT is 30 mm is φ = 2π × 5/30 = π / 3 [rad] from the equation (3). The correction coefficient φ = π / 3 is applied to the configuration shown in FIG. 1 to calculate the magnetic pole position θ.

In FIG. 5, the horizontal axis represents the actual magnetic pole position [mm], and the vertical axis represents the magnetic pole position error [mm]. In FIG. 5, reference numeral 35 indicates characteristics when the magnetic pole position detection is performed by the present technique using the configuration illustrated in FIG. 1, and reference numeral 36 indicates the magnetic pole position detection by the conventional technique under the condition of Expression (2). The characteristics of the case are shown.

As shown in FIG. 5, when the correction coefficient φ is set to π / 3, it can be seen that according to this method, the magnetic pole position error becomes zero and the magnetic pole position can be detected correctly. On the other hand, in the conventional method, it is understood that an error occurs in the magnetic pole position and a correct magnetic pole position cannot be obtained. This is because in the conventional method, the relationship between the magnetic pole pitch and the arrangement interval does not satisfy the constraint condition of Expression (2).

In this case, in order to correctly detect the magnetic pole position by the conventional method, it is necessary to set the arrangement interval L to 7.5 mm which is 1/4 of the magnetic pole pitch PIT. That is, by using this method, the magnetic pole position can be detected even when the arrangement interval is 5 mm shorter than 7.5 mm, so that the magnetic pole position detecting device 1 can be downsized.

(Example 2)
FIG. 6 is a diagram for explaining the magnetic pole position detection operation when the arrangement interval is 5 mm and the magnetic pole pitch is 20 mm in the configuration shown in FIG. That is, FIG. 6 shows a case where only the magnetic pole pitch is different from that of the first embodiment. FIG. 7 is a diagram showing a simulation result under the conditions shown in FIG. 6 in comparison with the conventional method.

The correction coefficient φ when the arrangement interval L is 5 mm and the magnetic pole pitch PIT is 20 mm is φ = 2π × 5/20 = π / 2 [rad] from the equation (3). The correction coefficient φ = π / 2 is applied to the configuration shown in FIG. 1 to calculate the magnetic pole position θ.

7, the horizontal axis represents the actual magnetic pole position [mm], and the vertical axis represents the magnetic pole position error [mm]. In FIG. 7, the code | symbol 37 shows the characteristic when the magnetic pole position detection is performed by the present technique using the configuration shown in FIG. 1 and when the magnetic pole position detection is performed by the conventional technique under the condition of Expression (2).

As shown in FIG. 7, according to this method, even when the correction coefficient φ is set to π / 2, the magnetic pole position error becomes zero, and the magnetic pole position can be detected correctly. On the other hand, the magnetic pole position can be correctly detected by the conventional method, and the magnetic pole position error is zero. This is because the condition shown in FIG. 6 is that the arrangement interval is 1/4 of the magnetic pole pitch and satisfies the constraint condition of the formula (2) used in the conventional method.

Thus, in this method, as shown in FIGS. 4 and 6, even if the magnetic pole pitch is different, the magnetic pole position error can be made zero by appropriately setting the correction coefficient φ. On the other hand, in the conventional method, the condition of FIG. 4 does not satisfy the constraint condition of Equation (2), so that the magnetic pole position cannot be calculated correctly, and a magnetic pole position error occurs.

Therefore, in this method, the magnetic pole position can be detected using a common magnetic pole position detection device by appropriately setting the correction coefficient φ even between linear motors having different magnetic pole pitches.

(Example 3)
FIG. 8 is a diagram showing a simulation result of the magnetic pole position obtained by calculating the correction coefficient when the arrangement interval is deviated from the theoretical value in the configuration shown in FIG. 2 in comparison with the conventional method. In FIG. 8, the theoretical value of the arrangement interval L is 5 mm shown in FIG. 6, but when the actual arrangement interval is 4.9 mm, the characteristic 38 and the formula ( A characteristic 39 when the magnetic pole position is detected from the sensor outputs a and b of the

magnetic sensors

11 and 12 by the conventional method under the condition 2) is shown. In FIG. 8, the horizontal axis represents the actual magnetic pole position [mm], and the vertical axis represents the magnetic pole position error [mm].

In the case of the conditions shown in FIG. 6, theoretically, even the conventional method satisfies the constraint condition of the expression (2), so that the magnetic pole position can be correctly detected as shown in FIG. Since the present example is a case where there is an error in the actual attachment, it corresponds to a case where the constraint condition of Expression (2) is not actually satisfied.

Therefore, when the magnetic pole position is calculated by the conventional method (2), the sensor output b is treated as cos θ even though the phase of the sensor output b is actually more advanced than the theoretical value. As shown in FIG. 39, an error of about 5.0-4.9 = 0.1 mm at maximum occurs.

In contrast, in this method, the actual arrangement interval is determined by actually measuring the arrangement interval of the magnetic sensor after installation or estimating the arrangement interval of the magnetic sensor after installation from the amplitude value of the magnetic pole position error waveform. When it is acquired, applied to Equation (3) to calculate the correction coefficient φ, and applied to the configuration of FIG. 1, errors due to attachment can be corrected. In this example, the attachment error is corrected by giving φ = 2π × 4.9 / 20 = 1.53938 to the correction coefficient φ, so that the magnetic pole position error becomes zero as shown in the characteristic 38. .

Thus, when the arrangement interval of the magnetic sensors is deviated from the theoretical value, it can be seen that the error is canceled by appropriately setting the correction coefficient φ by this method.

Example 4
FIG. 9 is a diagram for explaining a method of performing magnetic pole position detection using three magnetic sensors. When this method is used, since there is no restriction on the position of the magnetic sensor, the magnetic pole position can be detected using three or more magnetic sensors.

[Correction based on Rule 91 14.11.2008]
For example, as shown in FIG. 9, the third magnetic sensor 13 is added, the

magnetic sensors

11 and 12 are arranged at the position of the interval L1 = 2.5 mm, and the

magnetic sensors

12 and 13 are arranged at the interval L2 = 2.5 mm. Place in position. The magnetic pole position is detected by a combination of three magnetic sensors.

That is, the magnetic pole position θ12 is calculated using the magnetic sensor 11 and the magnetic sensor 12, the magnetic pole position θ13 is calculated using the magnetic sensor 11 and the magnetic sensor 13, and the magnetic pole position is calculated using the magnetic sensor 12 and the magnetic sensor 13. The position θ23 is calculated, and the magnetic pole position θ is
θ = (θ12 + θ13 + θ23) / 3
And calculate.

Thus, for example, since three magnetic sensors can be used in combination, the influence of noise or the like can be mitigated and the detection accuracy can be increased. Further, even if one of the three magnetic sensors fails, the magnetic pole position can be detected by the remaining two, so that the reliability of the magnetic pole detection device can be improved.

As described above, according to this embodiment, the phase difference φ calculated by the equation (3) that defines the relationship between the arrangement interval of the two magnetic sensors and the magnetic pole pitch (the number of poles in the rotary motor) is used as the correction coefficient. As described above, since each value of the two sensor outputs is corrected, a magnetic pole position detecting device without restriction between the magnetic pole pitch (the number of poles in the rotary motor) and the arrangement interval of the magnetic sensors can be realized.

In addition, the magnetic pole position detection device is configured to calculate and detect the magnetic pole position θ by using two sensor outputs and the phase difference φ as a correction coefficient as inputs, and the sine value of the phase difference φ is within the range of 1 >> δ> 0. Is larger than the determination threshold value δ determined in (5), the magnetic pole position θ is calculated and detected by the equation (5). When the sine sin φ of the phase difference φ is smaller than the determination threshold value δ, the magnetic pole position θ is calculated by the equation (6). Since it is configured to switch appropriately according to the magnitude of the sine value of the phase difference φ so as to detect the calculation, the magnetic pole position can be detected at an arbitrary arrangement interval of the two magnetic sensors.

Therefore, even if there is an error in the installation of the two magnetic sensors and the actual arrangement interval deviates from the theoretical value, the actual arrangement interval measured or the installation error is estimated from the amplitude value of the magnetic pole position error waveform. By using the correction coefficient φ calculated by applying the actual arrangement interval to Equation (3), it is possible to remove the attachment error. This eliminates the need for troublesome correction processing as in the prior art.

Also, there is no restriction due to the magnetic pole pitch (the number of poles in a rotary motor) as in the past, and the two magnetic sensors can be freely arranged at an arbitrary interval. Therefore, in the magnetic pole position detecting device mounted on the linear motor, it is possible to reduce the size of the device and to share the linear motor with different magnetic pole pitches.

In addition, since the degree of freedom of the arrangement of the magnetic sensor is increased, it is easy to mount three or more magnetic sensors in a magnetic pole position detection device for use in a linear motor, and the magnetic sensor has redundancy. Can do. As a result, the detection accuracy can be improved and the reliability can be improved.

As described above, the magnetic pole position detection device according to the present invention is useful as a magnetic pole position detection device that can correctly acquire the magnetic pole position even if two magnetic sensors are arranged at an arbitrary interval, and is particularly mounted on a linear motor. Suitable for use.

Claims

Of the two or more magnetic sensors arranged at a predetermined interval along the magnetic pole arrangement direction of the motor, the magnetic pole position where the movable part is located is calculated and detected based on the phase difference between the sensor outputs of the two magnetic sensors. In the magnetic pole position detection device, when the phase advance side sensor output of the two sensor outputs is a, the phase delay side sensor output is b, and the phase difference is φ,
The magnetic pole position θ when the sine value of the phase difference is larger than the determination threshold,
θ = tan −1 {(a · sinφ / (ba−cosφ)}
A first calculation system that detects by performing the calculation of
The magnetic pole position θ when the sine value of the phase difference is smaller than the determination threshold,
θ = {sin −1 (a) + sin −1 (b) −φ} / 2
A second arithmetic system that detects by performing the arithmetic operation;
A magnetic pole position detecting device comprising:
The first arithmetic system is:
A first multiplier that multiplies the phase advance sensor output of the two sensor outputs by the sine value of the phase difference, and a second multiplier that multiplies the phase advance sensor output by the cosine value of the phase difference. A subtractor, a first subtracter for subtracting the output value of the second multiplier from the slow phase sensor output of the two sensor outputs, and the output value of the first multiplier of the subtractor. A divider that divides by an output value; and a first angle calculator that calculates an arctangent value of a tangent value output by the divider and outputs it as the magnetic pole position;
The second arithmetic system is:
From the output value of the second angle calculator that calculates the inverse sine value of the phase advance side sensor output, the third angle calculator that calculates the inverse sine value of the phase delay side sensor output, and the third angle calculator A second subtractor for subtracting the phase difference; an adder for adding the output value of the second angle calculator and the output value of the second subtractor; A third multiplier that outputs a value obtained by multiplying by 2 as the magnetic pole position,
The magnetic pole position detection apparatus according to claim 1.
A phase difference calculation unit that obtains the phase difference by calculating (2π × interval between the two magnetic sensors) / (magnetic pole pitch or number of poles), and the arrangement interval is an error in the detected magnetic pole position. 3. The magnetic pole position detection device according to claim 1, wherein an arrangement interval actually measured or an arrangement interval estimated from an amplitude value of a magnetic pole position error waveform is applied.
The magnetic pole position according to any one of claims 1 to 3, wherein one or more magnetic sensors are added to the two magnetic sensors, and the magnetic pole position is detected by a combination of the plurality of magnetic sensors. Detection device.
The phase difference between the sensor outputs of two magnetic sensors out of two or more magnetic sensors arranged at a predetermined interval along the magnetic pole arrangement direction of the motor (2π × interval between the two magnetic sensors) / A first step of obtaining (magnetic pole pitch or number of poles) by calculation;
A second step of determining a magnitude relationship between the sine value of the phase difference and a determination threshold;
When the sine value of the phase difference is larger than the determination threshold value as a result of the determination in the second step, the phase difference sensor output of the sensor outputs of the two magnetic sensors is multiplied by the sine value of the phase difference. The tangent value with the value obtained by subtracting the value obtained by multiplying the phase advance sensor output by the cosine value of the phase difference from the sensor output of the two magnetic sensors as the denominator. A third step of calculating and detecting the arc tangent value of the first magnetic pole position as a first magnetic pole position;
As a result of the determination in the second step, when the sine value of the phase difference is smaller than a determination threshold, the sum of the reverse sine value of the phase advance side sensor output and the reverse sine value of the slow phase side sensor output A fourth step of calculating and detecting half the value obtained by subtracting the phase difference as the second magnetic pole position;
A magnetic pole position detection method comprising:
When there is an error in the calculated and detected magnetic pole position, the step of actually measuring the arrangement interval of the two magnetic sensors, or estimating from the amplitude value of the magnetic pole position error waveform;
Applying the actually measured arrangement interval or the estimated arrangement interval to the first step;
The magnetic pole position detection method according to claim 5, comprising: